A Longitudinal Cohort Study of Emergency Room Utilization Following Mifepristone Chemical and Surgical Abortions, 1999-2015

Deception by obfuscation: Studnicki et al.'s retracted longitudinal cohort study of emergency room utilization following abortion

OBJECTIVES

In November 2022, the anti-abortion advocacy group Alliance for Hippocratic Medicine filed a lawsuit against the U.S. Food and Drug Administration challenging the initial 2000 approval of mifepristone and its subsequent approvals, which removed unnecessary restrictions on its use, by disputing the medication's safety record. Such challenges relied on a study examining the incidence of emergency room visits following medication abortion with mifepristone and procedural abortion using Medicaid claims data from 1999-2015. In February 2024 that study was retracted by its publisher. In this paper, we analyzed the methods and presentations of the data used in the study.

STUDY DESIGN

We drew upon commonly accepted principles in responsible epidemiologic and scientific research to evaluate the methods and presentations of the data and organized our findings into themes.

RESULTS

We found multiple instances of methodological flaws, mischaracterizations, and obfuscations of data in this study, including use of a misleading research question and framing, analytic flaws, inappropriate use of an unvalidated proxy measure for outcomes of interest, and inappropriate and deceptive visualizations of data. In each instance, the resulting effect obfuscated and misrepresented the safety of medication abortion with mifepristone.

CONCLUSIONS

The misrepresentation and exaggeration of data promoted and exacerbated misinterpretations about the study's findings, resulting in substantial harm before it was retracted. Recognizing that ongoing judicial proceedings threaten access to conventional reproductive health care in the United States, public health policies must be informed by scientific and medical literature that is comprehensive, methodologically sound, and absent any obfuscations or misrepresentations.

IMPLICATIONS

Studnicki et al.'s study of emergency room visits after abortion misrepresented the safety of mifepristone with multiple instances of methodological flaws and obfuscations of data. While the study has now been retracted, it led to irrevocable harm, threatening access to medication abortion, which has an established safety record.

Self-Managed Medication Abortion: Implications for Clinical Practice

Medication abortion represents more than 50 percent of abortions in the United States (US). Since its approval in the US in 2000, the Food and Drug Administration (FDA) has progressively relaxed the prescribing requirements such that currently, no office visit, in-person dispensing, or ultrasound is required. Obtaining medication for abortion online without medical supervision or evaluation is also possible. This article reviews the complications of medication abortion by examining major studies and delineates the risks specific to self-managed abortion to inform clinicians in caring for women.

Summary

Medication abortion has become the most common abortion method in the United States. This document provides a detailed history of the relaxation requirements on medication abortion and reviews the major studies on medication abortion complications including a discussion of their limitations. Finally, the paper delineates the ease of access to medication abortion without a health care provider and the risks associated with self-managed abortion. This paper is intended to provide information for clinicians who likely will be encountering increasing number of patients with such complications.

Short-Term Adverse Outcomes After Mifepristone-Misoprostol Versus Procedural Induced Abortion : A Population-Based Propensity-Weighted Study

BACKGROUND

Prior studies comparing first-trimester pharmaceutical induced abortion (IA) with procedural IA were prone to selection bias, were underpowered to assess serious adverse events (SAEs), and did not account for confounding by indication. Starting in 2017, mifepristone-misoprostol was dispensed at no cost in outpatient pharmacies across Ontario, Canada.

OBJECTIVE

To compare short-term risk for adverse outcomes after early IA by mifepristone-misoprostol versus by procedural IA.

DESIGN

Population-based cohort study.

SETTING

Ontario, Canada.

PATIENTS

All women who had first-trimester IA.

MEASUREMENTS

A total of 39 856 women dispensed mifepristone-misoprostol as outpatients were compared with 65 176 women undergoing procedural IA at 14 weeks' gestation or earlier within nonhospital outpatient clinics (comparison 1). A total of 39 856 women prescribed mifepristone-misoprostol were compared with 8861 women undergoing ambulatory hospital-based procedural IA at an estimated 9 weeks' gestation or less (comparison 2). The primary composite outcome was any SAE within 42 days after IA, including severe maternal morbidity, end-organ damage, intensive care unit admission, or death. A coprimary broader outcome comprised any SAE, hemorrhage, retained products of conception, infection, or transfusion. Stabilized inverse probability of treatment weighting accounted for confounding between exposure groups.

RESULTS

Mean age at IA was about 29 years (SD, 7); 33% were primigravidae. Six percent resided in rural areas, and 25% resided in low-income neighborhoods. In comparison 1, SAEs occurred among 133 women after mifepristone-misoprostol IA (3.3 per 1000) versus 114 after procedural IA (1.8 per 1000) (relative risk [RR], 1.87 [95% CI, 1.44 to 2.43]; absolute risk difference [ARD], 1.5 per 1000 [CI, 0.9 to 2.2]). The respective rates of any adverse event were 28.9 versus 12.4 per 1000 (RR, 2.33 [CI, 2.11 to 2.57]; ARD, 16.5 per 1000 [CI, 14.5 to 18.4]). In comparison 2, SAEs occurred among 133 (3.4 per 1000) and 27 (3.3 per 1000) women, respectively (RR, 1.04 [CI, 0.61 to 1.78]). The respective rates of any adverse event were 31.2 versus 24.9 per 1000 (RR, 1.25 [CI, 1.04 to 1.51]).

LIMITATION

A woman prescribed mifepristone-misoprostol may not have taken the medication, and the exact gestational age at IA was not always known.

CONCLUSION

Although rare, short-term adverse events are more likely after mifepristone-misoprostol IA than procedural IA, especially for less serious adverse outcomes.

PRIMARY FUNDING SOURCE

Canadian Institutes of Health Research.

A Post Hoc Exploratory Analysis: Induced Abortion Complications Mistaken for Miscarriage in the Emergency Room are a Risk Factor for Hospitalization

Introduction

Previous research indicates that an increasing number of women who go to an emergency room for complications following an induced abortion are treated for a miscarriage, meaning their abortion is miscoded or concealed.

Objective

To determine if the failure to identify a prior induced abortion during an ER visit is a risk factor for higher rates of subsequent hospitalization.

Methods

Post hoc analysis of hospital admissions following an induced abortion and ER visit within 30 days: 4273 following surgical abortion and 408 following chemical abortion; abortion not miscoded versus miscoded or concealed at prior ER visit.

Results

Chemical abortion patients whose abortions are misclassified as miscarriages during an ER visit subsequently experience on average 3.2 hospital admissions within 30 days. 86% of the patients ultimately have surgical removal of retained products of conception (RPOC). Chemical abortions are more likely than surgical abortions (OR 1.80, CL 1.38-2.35) to result in an RPOC admission, and chemical abortions concealed are more likely to result (OR 2.18, CL 1.65-2.88) in a subsequent RPOC admission than abortions without miscoding. Surgical abortions miscoded/concealed are similarly twice as likely to result in hospital admission than those without miscoding.

Conclusion

Patient concealment and/or physician failure to identify a prior abortion during an ER visit is a significant risk factor for a subsequent hospital admission. Patients and ER personnel should be made aware of this risk.